Re: [Flightgear-devel] [FlightGear-devel] amazing FDM initialization bug

2008-12-31 Thread Jon S. Berndt
John Denker wrote: > I've always thought the 200-ft business was a weird optimization, > especially since other properties such as gear-extension-norm > are recomputed at frame rate at all altitudes. And that is > the solution for retractable-gear airplanes: forget about wow > and just look at g

Re: [Flightgear-devel] amazing FDM initialization bug

2008-12-31 Thread Ron Jensen
On Wed, 2008-12-31 at 20:07 -0700, John Denker wrote: > On 12/31/2008 06:22 PM, Ron Jensen wrote: > > > for what its worth, the in-air dialog could > > set /fdm/jsbsim/gear/unit[*]/WOW to false. > > I thought of that. > > It doesn't work. Yes, it does. > The FDM rewrites the "wow" property a

Re: [Flightgear-devel] amazing FDM initialization bug

2008-12-31 Thread John Denker
On 12/31/2008 06:22 PM, Ron Jensen wrote: > for what its worth, the in-air dialog could > set /fdm/jsbsim/gear/unit[*]/WOW to false. I thought of that. It doesn't work. The FDM rewrites the "wow" property at frame rate, forcing it to 1. Why it is so insistent on writing a value to the pr

Re: [Flightgear-devel] amazing FDM initialization bug

2008-12-31 Thread Ron Jensen
On Wed, 2008-12-31 at 12:20 -0600, Curtis Olson wrote: > The way this was explained to me is that JSBSim only performs the wow > test when it is within 200' of the ground. Unfortunately, this means > that if you start in the air higher than that, the gear variables can > be left in an unsettled st

Re: [Flightgear-devel] amazing FDM initialization bug

2008-12-31 Thread Curtis Olson
On Wed, Dec 31, 2008 at 12:35 PM, John Denker wrote: > On 12/31/2008 11:20 AM, Curtis Olson wrote: > > The way this was explained to me is that JSBSim only performs the wow > test > > when it is within 200' of the ground. Unfortunately, this means that if > you > > start in the air higher than t

Re: [Flightgear-devel] amazing FDM initialization bug

2008-12-31 Thread John Denker
On 12/31/2008 11:20 AM, Curtis Olson wrote: > The way this was explained to me is that JSBSim only performs the wow test > when it is within 200' of the ground. Unfortunately, this means that if you > start in the air higher than that, the gear variables can be left in an > unsettled state because

Re: [Flightgear-devel] amazing FDM initialization bug

2008-12-31 Thread Curtis Olson
The way this was explained to me is that JSBSim only performs the wow test when it is within 200' of the ground. Unfortunately, this means that if you start in the air higher than that, the gear variables can be left in an unsettled state because the test that sets the variables never gets run. R

Re: [Flightgear-devel] amazing FDM initialization bug

2008-12-31 Thread John Denker
I have a pretty-much workable workaround for the worst of these bugs. The big clue is that "presets-commit" apparently just deletes the old FDM instance and creates another. To make the new FDM happy, I had to delete a bunch of nodes from the property tree. Some other nodes needed to be set to -

[Flightgear-devel] amazing FDM initialization bug

2008-12-29 Thread John Denker
I've been trying to use the presets-commit feature ... with less-than-satisfactory results. All I wanted to do is pop the aircraft into the air, with a reasonable airspeed, attitude, et cetera. Note that I set presets/trim {BOOL} = 0 since the trim was known to be already OK, and be